Analysis of meniscal

Transcription

1 Knee Surg Sports Traumatol Arthrosc (2004) 12 : KNEE DOI /s z Reha N. Tandogan Ömer Taşer Asım Kayaalp Emin Taşkıran Halit Pınar Bülent Alparslan Aziz Alturfan Analysis of meniscal and chondral lesions accompanying anterior cruciate ligament tears: relationship with age, time from injury, and level of sport Received: 3 October 2001 Accepted: 7 April 2003 Published online: 20 September 2003 Springer-Verlag 2003 This multi-center study was performed by the Turkish Society of Sports Traumatology, Arthroscopy and Knee Surgery. It was presented as a poster at the ESSKA 2000 Congress in London, UK R. N. Tandogan ( ) and Traumatology, Başkent University, Fevzi Çakmak caddesi 10.sokak no.45, Bahçelievler, Ankara, Turkey Tel.: , Fax: , rehat@baskent-ank.edu.tr Ö. Taşer A. Alturfan and Traumatology, Istanbul University, Topkapı, İstanbul, Turkey A. Kayaalp Clinic of Orthopedics and Traumatology, Çankaya Hospital, Kavaklıdere, Ankara, Turkey E. Taşkıran and Traumatology, Ege University, Bornova, İzmir, Turkey H. Pınar and Traumatology, Dokuz Eylül University, Balçova, İzmir, Turkey B. Alparslan and Traumatology, Adnan Menderes University, Aydın, Turkey Abstract The purposes of this multi-center study were: (a) to document the location and type of meniscal and chondral lesions that accompany anterior cruciate ligament (ACL) tears, and (b) to test for possible relationships between these lesions and patient age, time from initial injury (TFI), and sports level (i.e., recreation, amateur, professional, and national). The cases of 764 patients with ACL tears who underwent arthroscopy for the first time were retrospectively analyzed. The group included 684 males and 80 females of mean age 27 years (range years). The mean TFI was 19.8 months (range months). Eighty-seven percent of the group engaged in regular sporting activity. Thirty-seven percent had medial meniscal tears, 16% had lateral meniscal tears, and 20% had tears of both menisci. The most common tear types were longitudinal tears in the posterior and middle horns of both menisci. Tears of the lateral meniscus were more centrally located than those of the medial meniscus. Incomplete tears and radial tears were significantly more common in the lateral meniscus. Nineteen percent of the knees had one or more chondral lesions. Sixty percent of the chondral lesions were located in the medial tibio-femoral compartment. Patient age was statistically associated with presence of a medial meniscal tear, presence of a grade 3 or 4 chondral lesion, and presence of a complex tear of the medial meniscus. Sports level was not statistically related to any of the parameters studied. The odds of having a medial meniscal tear at 2 to 5 years TFI were 2.2 times higher than the odds in the first year post-injury, and the odds at >5 years were 5.9 times higher than at 0 to 12 months TFI. The frequency of lateral meniscal tear remained fairly constant at 2 years TFI. The odds of having a grade 3 or 4 chondral lesion were 2.7 times greater at 2 to 5 years TFI than they were at 1 year post-injury, and these odds increased to 4.7 when patients at >5 years TFI were compared to those in the 2 to 5 years category. Multivariate analysis demonstrated that TFI and age were equally important predictors of lateral meniscal tears and of grade 3 or 4 chondral lesions; however, TFI was the better predictor of medial meniscal tear. Keywords Anterior cruciate ligament Meniscus tear Cartilage damage Sports injury

2 263 Introduction Anterior cruciate ligament (ACL) injuries are common in sports, and the meniscal tears that accompany this type of damage have been explored extensively in the literature. Several studies have revealed a preponderance of lateral meniscal tears in the acute setting, and a rise in the incidence of medial tears as the time from injury increases [3, 16]. Others have found medial meniscal tears to be more common in the acute setting [2, 12, 17]. Keene et al. [12] have shown that the complexity of the tear increases in the chronic stage, and tears are less amenable to repair as time passes. Long-term follow-up studies have demonstrated that patients with ACL injuries are at increased risk for early gonarthrosis, even if knee stability is restored after ACL reconstruction [5, 7, 14, 19]. The relationship between meniscectomy and arthritis has been well documented; however, chondral lesions that occur as part of the initial injury, and subsequent episodes of knee subluxation, may also contribute to the development of arthritis. The purpose of this study was twofold. Our first objective was to document the location and type of meniscal and chondral lesions that accompany ACL tears in a large patient population. Our second aim was to test for relationships between the above-mentioned lesions and the following variables: 1. Patient age 2. Time from initial injury (TFI) 3. Sports level Patients and methods This retrospective multi-center study involved patients who were operated on by experienced instructors of the Turkish Society of Sports Traumatology, Arthroscopy, and Knee Surgery. All surgeons worked in tertiary care referral centers and had at least 10 years of arthroscopic surgery experience at the time of the study. Seven hundred and sixty-four patients with ACL tears who were undergoing arthroscopy for the first time were included. Subjects were excluded if they had had previous surgery, had other ligamentous lesions, or if their records were incomplete. Surgeons from six centers filed a standard evaluation form for each case. Sports level was defined as: recreational (participation 2 3 times/week, no official tie with a team); amateur (member of a second or third division team); professional (member of a first division team); or national (elite athlete on the national team). The time of the initial ACL injury was determined by the patient s history of injury, with special emphasis on the presence of major twisting injury, the knee giving way with a popping sound, hemarthrosis, or inability to return to sport or walking. The nature of this injury was verified via the hospital records or the records of the primary treating physician where available. Patients without a clear history of a traumatic episode were excluded. Since several years had passed since the time of injury in some patients, TFI was measured in months instead of days to minimize errors. In addition to patient demographics, data on the following parameters were collected: 1. Meniscal tears Location (according to zones described by Cooper et al. [4]) Extent (full-thickness or partial-thickness) Type (longitudinal, flap, radial, complex) 2. Chondral lesions Number of lesions Location, grade, and size, according to the Cartilage Standard Evaluation Form/Knee set out by the ICRS (International Cartilage Repair Society) [8]. The sizes of the lesions were measured with 3 mm calibrated probes. The five most common tear locations were mapped according to the zones described by Cooper et al. The frequencies of chondral lesions were also mapped on a chart proposed by the ICRS. We statistically analyzed how patient age, TFI, and sports level were related to the following lesion parameters: presence of a meniscal tear; type of meniscal tear; extent of meniscal tear; number of chondral lesions; and presence of chondral lesions higher than grade 3. Data analysis was performed using the software SPSS v. 7.0 for Windows. Analysis of variance (ANOVA), Student s t-test, and Chi-square testing were used for the elementary analysis. Crude odds ratios were computed for statistically significant differences where applicable. The resultant odds ratios were also subjected to multivariate analysis (logistic regression) to assess for confounding factors. The Pearson correlation coefficient was calculated to analyze the relationship between TFI and the number of chondral lesions. Results Patient demographics There were 684 (89.5%) male and 80 (10.5%) female patients. The mean age in the group was 27±7 years (range years). The mean time from injury (TFI) was 19.8± 32.7 months (range months). Six hundred and sixty-three (86.8%) of the patients engaged in regular sporting activity. The types and levels of sports participation are listed in Table 1. More than 65% of the patients were recreational soccer players. Table 1 The distribution of types and levels of sports participation a Other sports: tennis, wrestling, karate, gymnastics, track, cycling, boxing, professional dancing, parachuting, tae kwon do, table tennis Type of sport Recreational Amateur Professional National Total Soccer Basketball Volleyball Handball Skiing Other a Total

3 264 Meniscal lesions A total of 556 (72.8%) patients had one or more meniscal tears. Two hundred and seventy-nine (36.5%) of the cases had medial meniscal tears, 121 (15.8%) had lateral meniscal tears, and 156 (20.4%) had torn both the medial and lateral menisci. Thirty-nine (5.1%) of the patients had a second tear in the same meniscus. Twenty-six (3.4%) of these were in the medial meniscus, and 13 (1.7%) were in the lateral meniscus. These second tears are included in the tables that summarize tear location, type, and extent, but were not statistically analyzed as a separate category in relation to patient age, TFI, and sports level. Figure 1 illustrates the five most common medial and lateral meniscal tears, mapped according to Cooper s zones. The most frequent tear locations on the medial side were A2B2, A1B1, A1, A2, and A2B2C2. These corresponded to longitudinal tears in the posterior and middle onethirds. The five most common tear sites on the lateral side were E2F2, F2, F3, E3F3, and F1. These were also in the posterior and middle one-thirds, but tended to be more centrally located in contrast to tears on the medial side. Thirty-one of the 461 medial meniscal tears were partial-thickness (6.7%) and 48 of the 290 lateral meniscal tears were partial-thickness (16.6%). This difference was statistically significant (χ 2 =18.26, p=0.000). Figure 2 shows the distribution of tear types. Seventyseven percent of the medial and 64% of the lateral meniscal tears were longitudinal. Radial tears were more common on the lateral side than the medial (15.8% vs 1.5%, respectively), and this difference was statistically significant (χ2=55.84, p=0.000). Chondral lesions One hundred forty-six patients (19.1%) had one or more chondral lesions. Eighty patients had a single lesion, 41 had two, 9 had three, 10 had four, 5 had five, and 1 had six separate chondral lesions in the knee. Figure 3 shows the distribution of these lesions according to a modified version of the ICRS classification. Forty-one percent of the chondral lesions were in the middle part of the medial femoral condyle. Sixty percent of the chondral lesions were located in the medial tibiofemoral compartment. The ICRS Cartilage Standard Evaluation Form was used to grade these lesions. Sixty-one (23%) were grade 1, 113 (44%) were grade 2, 58 (22%) were grade 3, and 28 (11%) were grade 4 chondral lesions. The mean area of these lesions was 219±175 mm 2 (range 10 mm mm 2 ). Patient age Age vs presence of meniscal tear Fig. 1 A map of the five most common types of meniscal tears, identified according to Cooper s zones The mean age of the patients with medial meniscal tears was 28.0±7.6 years, whereas the mean age of those who had no meniscal tearing was 25.4±6.7. This difference was statistically significant (Student s t-test, p=0.000). In contrast, the mean ages of patients with and without lateral meniscal tears were 27.8±7.4 years and 26.4±7.3 years respectively. This difference was not statistically significant (Student s t-test, p=0.09). Age vs type of tear Fig. 2a, b The distribution of meniscal tear types. a Medial meniscal types; b lateral meniscal types ANOVA showed that there was a statistically significant relationship between medial meniscal tear type and patient age (p=0.0026). Student s t-tests were performed to compare the mean ages in the groups with different tear types. These results are summarized in Table 2. The analysis revealed that patients with a complex tear in the medial meniscus were approximately 3 years older than those who had longitudinal or flap tears. ANOVA revealed no statistically significant relationship between lateral meniscal tear type and patient age (p=0.751).

4 265 Fig. 3 The distribution of chondral lesions according to a modified version of the International Cartilage Research Society (ICRS) classification. The numbers indicate the total number of lesions seen at each given location Table 2 Mean ages of patients with different medial meniscal tear types Medial meniscus tear Age vs extent of meniscal tear There was no statistically significant relationship between extent of meniscal tear (full- or partial-thickness) and patient age, based on ANOVA (p=0.351 for medial meniscus, p=0.290 for lateral meniscus). Age vs number of chondral lesions Mean age Longitudinal 27.4±7.6* Flap 27.7±7.5* Radial 30.7±6.8 Complex 31.4±6.7* *Difference between longitudinal and complex tears statistically significant (p=0.000); difference between flap and complex tears statistically significant (p=0.022); no other statistically significant relationships ANOVA showed no statistically significant relationship between patient age and number of chondral lesions (p=0.155). Age vs presence of a grade 3 or 4 chondral lesion The mean age of patients with grade 3 or 4 chondral lesions was 31.8±8.5 years, whereas the mean age of those with no chondral lesions or grade 1 or 2 lesions was 26.2±6.9 years. This difference was statistically significant (Student s t-test, p=0.000). Sports level There was no statistically significant relationship between level of sports participation and presence of a medial meniscal tear (χ 2 =4.5, p=0.205), presence of a lateral meniscal tear (χ 2 =6.11, p=0.106), medial meniscal tear type (χ 2 =7.8, p=0.551), lateral meniscal tear type (χ 2 =11.02, p=0.274), number of chondral lesions in the knee (ANOVA, p=0.932), or presence of grade 3 or 4 chondral lesions (χ 2 =4.71, p=0.194). Time from initial injury TFI vs presence of a meniscal tear The mean TFI in patients with a medial meniscal tear was 26.1±39.3 months, whilst that for patients who had no

5 266 Fig. 4 The percentages of patients with a medial and/or lateral meniscal tear versus TFI Table 3 Odds ratios (OR) for the presence of a medial and lateral meniscal tear in different categories of TFI *Statistically significant Time from Medial meniscus tear+ Lateral meniscus tear+ initial injury (months) OR Confidence interval P value OR Confidence interval P value 0 12 vs <OR< * <OR< * 0 12 vs <OR< * <OR< * vs <OR< * <OR< medial meniscal tearing was 11.4±17.8 months. This difference was statistically significant (Student s t-test, p= 0.000). The mean TFI in individuals who had lateral meniscal tears was 25.5±41.2 months, whilst the mean TFI for those with no lateral meniscal tearing was 16.6± 26.2 months. This difference was also statistically significant (Student s t-test, p=0.000). The proportions of patients who had meniscal tears at different time intervals are illustrated in Fig. 4. The percentage of patients with either a medial or lateral meniscal tear increased in parallel with TFI. Eighty-six percent of the patients with TFI >5 years had a medial meniscal tear. The corresponding figure for lateral meniscal tear was 48.3%. We also categorized the patients in three TFI groups (<1 year, 2 5 years, and >5 years) and then calculated each group s crude odds ratio for presence of a meniscal tear. The results are shown in Table 3. Patients at 2 5 years TFI were 2.2 times more likely to have a medial meniscal tear than those in the first year TFI. The odds of having such a tear at >5 years TFI were 5.9 times greater than the odds in the first year post-injury. The mean TFI of patients with flap tears was 28.8 months. There was a notable difference between TFI in patients with flap versus complex tears, but this was not statistically significant (p=0.057). We found no other statistically significant links between TFI and tear type. TFI vs extent of tear ANOVA testing showed no statistical association between TFI and extent of medial (p=0.085) or lateral (p=0.311) meniscal tearing. TFI vs number of chondral lesions There was a significant relationship between TFI and the number of chondral lesions in the knee (p=0.000), according to ANOVA. The number of chondral lesions increased with longer TFI. The Pearson correlation coefficient for TFI vs number of chondral lesions was 0.47, signifying a moderate relationship. TFI vs type of tear ANOVA revealed no statistically significant relationship between the type of lateral meniscal tear and TFI (p= 0.107), but there was a significant relationship between TFI and medial meniscal tear type (p=0.000). The mean TFI of patients with complex tears was 54.9 months, while mean TFI for those with longitudinal tears was 21.2 months. This difference was statistically significant (p=0.000). TFI vs grade 3 or 4 chondral lesions Table 4 lists the distribution of patients with grade 3 or 4 chondral lesions according to TFI interval. There were statistically significant differences among the numbers of patients with grade 3 or 4 chondral lesions in each of the three TFI intervals (χ 2 =81.64, p=0.000). The crude odds ratios for patients in each of three TFI categories with these grades of chondral lesion are listed in Table 5.

6 267 Table 4 The distribution of patients with grade 3 or 4 chondral lesions, listed according to TFI Time from initial injury (months) Total Patients with grade 3 or 4 chondral lesions Number of patients in each TFI interval Percentage 6.1% 14.8% 44.8% 11.3% Table 5 Odds ratios for presence of a grade 3 or 4 chondral lesion in the three TFI intervals Time from initial injury (months) Multivariate analysis Since both TFI and patient age seemed to be related to the presence of a meniscal tear and presence of a grade 3 or 4 chondral lesion, logistic regression analysis was performed to adjust for confounding factors. This revealed that TFI was the better predictor of medial meniscal tear, and the adjusted odds ratio for this was 1.1 (range ). TFI and age were equally good predictors of lateral meniscal tear, and the respective adjusted odds ratios were 1.06 (range ) and (range ). TFI and age were also similar predictors of a grade 3 or 4 chondral lesion, and the respective adjusted odds ratios in this case were 0.97 (range ) and 0.95 (range ). Discussion Grade 3 or 4 chondral lesion Odds Confidence P value ratio (OR) interval 0 12 vs <OR< * 0 12 vs <OR< * vs <OR< * *Statistically significant The changes that occur after untreated ACL have been extensively documented. Although Bonamo et al.[1] could not demonstrate the effect of patient age, associated meniscal and chondral injury on the outcome of 79 patients with conservatively treated ACL injury, most studies suggest that the incidence of meniscal tears rises with longer follow-up, and that subsequent meniscectomy leads to early radiological arthritis [14, 19, 20]. Gillquist and Messner [7], in their review of the literature, found that the risk of gonarthrosis increased tenfold in isolated ACL ruptures, and this risk further increased in combined ACL and meniscus injuries. The two objectives of this study were to document the meniscal and chondral lesions that accompany ACL tears in a large patient population, and to identify possible relationships between these lesions and patient age, time from initial injury, and level of sports participation. More than 70% of the patients had a meniscal tear at the time of arthroscopy. The most common tear types were longitudinal tears in the posterior and middle onethirds of both menisci. Interestingly, tears of the lateral meniscus were more centrally located than medial meniscal tears. Smith at al. [21] have also reported similar findings in their series of 476 ACL deficient patients, in which 75% of the medial meniscal tears were peripheral, whilst this ratio was 44% for lateral tears. Partial-thickness tears and radial tears were significantly more common in the lateral meniscus. The relative frequency of partial thickness and radial tears in the lateral meniscus has also been reported in previous studies [2, 6]. Several published studies on ACL injuries report a preponderance of lateral meniscal tears in the acute setting, and an increase of medial meniscal tears as time passes. In a series of 333 patients with ACL tears, Cipolla et al. [3] demonstrated lateral meniscal involvement in 51 % and medial meniscal injury in 31% of the patients in the acute stage (less than 2 weeks from injury). Lateral involvement was 48% in the chronic cases, but medial meniscal involvement rose to 70% in their series. Nikolic [16] reported that the most frequent finding associated with an acute ACL tear in 66 patients was a lateral meniscal tear, with an incidence of 72%. He could demonstrate no statistical difference between the incidence of a lateral meniscal tear in acute and chronic cases. In contrast, Paletta et al. [17] reported on 150 patients with acute ACL injuries and found that there was a strong trend of increased incidence of medial meniscal tears in non-skiers compared to skiers. Keene et al.[12] reviewed the meniscal status of 176 consecutive patients undergoing ACL reconstruction. The most common tear was the single vertical longitudinal split of the medial meniscus. Cerabona et al. [2] also reported that the most common tear type associated with an acute ACL tear was a peripheral longitudinal tear in the medial meniscus. In our series, there was a higher frequency of medial meniscal tears than lateral tears in all the time periods after injury. It can be postulated that the differences in the frequency of medial and lateral tears may be due to the different mechanisms of injury, lower extremity alignment and timing of surgery in different patient series, but further studies are necessary to clarify this issue. We used the ICRS Cartilage Standard Evaluation Form [8] to document the chondral lesions in the knee, since this allowed us to gather detailed information about the location, grade, and size of the lesion(s). Nineteen percent of the patients had one or more chondral lesions in their knees,

7 268 and the sizes of these ranged from 10 mm 2 to 900 mm 2. Forty percent of the lesions were in the middle part of the medial femoral condyle, and 60% involved the medial tibio-femoral compartment. Nearly 80% of the chondral lesions were associated with a meniscal tear in the same compartment. Other common locations were the middle part of the lateral femoral condyle, the middle part of the lateral tibial plateau, and the central-middle part of the patella. In a recent study, Murell et al. [15] reported on 130 patients undergoing ACL reconstruction. They found that the largest area of full thickness cartilage damage in ACL deficient knees was the medial femoral condyle. There was a positive correlation between meniscal loss and cartilage damage. In our study, the patients who had medial meniscal tearing were older than those who did not exhibit this type of damage. We noted no such relationship between patient age and presence of a lateral meniscal tear. The frequency of complex medial meniscal tears increased significantly with advancing age. Patient age was also related to the severity of chondral lesions. The patients with grade 3 or 4 chondral lesions were approximately 5 years older than those who had no chondral lesions, and this difference was statistically significant. Roos et al. [19] studied the influence of patient age and time from injury on osteoarthritic changes in the knee in 1012 patients with ACL and/or meniscal deficiency. Using arthroscopy and weight-bearing X-rays, they found that the first radiological changes were seen on average 10 years after the injury. They also noted that arthritic changes in patients with meniscal injury were seen about 10 years earlier in patients over 30 years of age. They concluded that arthritic changes occurred sooner in older patients with knee injury. Sports level was not statistically related to any of the parameters we studied. Since more than half the patients played soccer, we did an analysis to determine whether the soccer players sustained different injuries than patients who engaged in other sports. No such difference was demonstrated. Roos et al. [18] compared 71 former elite soccer players with non-elite players and age matched controls. They found that the risk of gonarthrosis was nearly 3 times higher in elite players, even in the absence known injury. Larsen et al. [13] reached similar conclusions in their review of 69 former elite soccer players. It seems that although the initial injury patterns are not related to sport level, long term sequela of knee injury are more significant in elite players. Our study revealed that TFI was associated with many of the lesion parameters. The frequency of medial meniscal injuries and lateral meniscal injuries rose as TFI increased. The odds of having a medial meniscal tear were 2.2 times higher at 2 to 5 years post-injury than in the first year post-injury, and were 5.9 times higher when TFI was greater than 5 years. The corresponding odds ratios for lateral meniscal tears for the same time points were 1.5 and 1.9. These findings indicate that the incidence of lateral meniscal tears changes only minimally after 2 years, whereas the incidence of medial meniscal tears continues to rise steadily as TFI increases. These results are similar to the findings of study by Keene et al. [12], who reported that there was a statistically significant increase in the incidence of medial meniscal tears as the injury became more chronic in 176 patients undergoing ACL reconstruction. Lateral meniscal tears remained fairly constant in their series. We also noted a relationship between TFI and type of medial meniscal tear. The incidence of complex tears increased with TFI in our series. Keene et al. [12] reported that the incidence of repairable medial meniscal tears was 80% in the acute setting. This ratio fell to 46% in the chronic stage, indicating that meniscal tears became less amenable to suturing in chronic cases. As expected, there was a significant association between TFI and the number of chondral lesions in the knee; however, the strength of the correlation was only moderate (r=0.47). TFI was also specifically related to the presence of grade 3 or 4 chondral lesions in the knee. The odds of having a serious chondral lesion were 2.7 times higher at 2 to 5 years post-injury than at 1 year TFI. The most dramatic finding was that patients at >5 years TFI were 12.6 times more likely to have chondral lesions than individuals at 1 year post-injury. Since both patient age and TFI seemed to be strongly related to the presence of a meniscal tear and/or presence of a serious chondral lesion, we performed multivariate analysis (logistic regression) to adjust for confounding factors. This yielded some interesting results. The adjusted odds ratios were much lower, and, in some cases, were only marginally significant. TFI and age were equally good predictors of lateral meniscal tearing and of grade 3 or 4 chondral lesions, but TFI was the better predictor of medial meniscal tear. Murrell et al. [15] also conducted a multivariate analysis of the relationship between cartilage damage and time from injury or age. They found that full thickness cartilage loss could be predicted from a linear combination of age and days from injury. A recent study found a 20% inter-observer variation in the interpretation of arthroscopic findings between trainees and experienced surgeons [10]. This study questioned the validity of results of multi-centric studies conducted by surgeons of different levels of experience. As our study was carried out in 6 different centers, it has all the weaknesses of a multi-centric study. Inter-observer differences in the evaluation and documentation of arthroscopic findings cannot be overruled completely. However, as all participating surgeons were experienced instructors of the Turkish Society of Knee Surgery Arthroscopy and Sports Traumatology working in tertiary care centers, the risk of a misdiagnosis was minimized. Van Kampen et al. [22] reported on the interobserver variance in diagnostic arthroscopy of 3 videotaped examinations, made by 91 experienced arthroscopists, and found little correlation between

8 269 the examiners in evaluating pathologic changes. However, the diagnoses were made solely by viewing the videotape, thus probing of the pathological lesion, which is an integral part of diagnosis, could not be carried out. This might be one of the reasons for the low agreement rates between surgeons. The Cooper Map and the ICRS Classification Systems used in our study are visually oriented and descriptive instruments that allow a detailed and reproducible documentation of arthroscopic findings, which we believe minimizes errors in documentation. The two other weaknesses of our study were: (1) it was not a longitudinal one, and (2) only symptomatic cases were analyzed. On this basis, our study population may not necessarily be representative of all patients with ACL tears. In the strict sense, our findings cannot be extrapolated to patients with ACL tears who have meniscal or chondral lesions but do not seek medical attention after injury, or to those who do not exhibit symptoms. However, since our study group included large numbers of patients in each of the TFI intervals, these results can be viewed as a reasonable overview of the meniscal and chondral lesions that accompany ACL injury. Several studies have evaluated the effect of meniscal preservation on the prevention of osteoarthritis in ACL deficient knees. Jomha et al. [11] reported on 53 patients with 7 year radiographic follow-up. Patients with early reconstruction of ACL and intact menisci had the best results, with 17 of the 20 patients having normal radiographs. In contrast, none of the 16 patients undergoing late ACL reconstruction and meniscectomy had normal radiographs. However, meniscal preservation may not be the only factor in the prevention of osteoarthritis. Jager et al. [9] reported on the 6 year follow-up of 50 sportsmen with ACL deficient knees. Repair of the torn menisci without reconstruction of the ACL could not prevent osteoarthritic changes in the knee. It seems that both restoration of knee stability and preservation of menisci are important to minimize osteoarthritis. In summary, we found that the frequency of medial meniscal tears and of chondral lesions increases with age and with time from initial injury in patients with ACL deficient knees. Lateral meniscal tearing seems to level off after 2 years post-injury. More than 60% of the chondral lesions were located in the medial tibio-femoral compartment, predominantly on the femoral condyle. TFI and patient age were equally important predictors of lateral meniscal tears and of grade 3 or 4 chondral lesions; however, TFI was the better predictor of medial meniscal tear. Acknowledgement The authors wish to thank Associate Professor Seval Akgün, Department of Public Health, Başkent University for her invaluable assistance with the statistical analysis. References 1. Bonamo JJ, Fay C, Firestone T (1990) The conservative treatment of anterior cruciate deficient knee. Am J Sports Med 18: Cerabona F, Sherman MF, Bonamo JR, Sklar J (1988) Patterns of meniscal injury with acute anterior cruciate ligament tears. Am J Sports Med 16: Cipolla M, Scala A, Gianni E, Puddu G (1995) Different patterns of meniscal tears in acute anterior cruciate ligament (ACL) ruptures and in chronic ACL-deficient knees. Classification, staging and timing of treatment. Knee Surg Sports Traumatol Arthrosc 3: Cooper DE, Arnoczky SP, Warren RF (1990) Arthroscopic meniscal repair. Clin Sports Med 9: Daniel DM (1993) Selecting patients for ACL surgery. In: Jackson DW (ed) The anterior cruciate ligament. Current and future concepts. Raven Press, New York, pp Fitzgibbons RE, Shelbourne KD (1995) Aggressive nontreatment of lateral meniscal tears seen during anterior cruciate ligament reconstruction. Am J Sports Med 23: Gillquist J, Messner K (1999) Anterior cruciate ligament reconstruction and the long-term incidence of gonarthrosis. Sports Med 27: International Cartilage Repair Society (1998) The Cartilage Standard Evaluation Form/Knee. ICRS Newsletter Spring 1998, p 7 9. Jager A, Braune C, Welsch F, Khoudeir S, Rauschmann MA (2002) Is meniscal repair an adequate procedure to prevent early osteoarthritis in athletes with chronic anterior knee instability? Zentralbl Chir 127: Javed A, Siddique M, Vaghela M, Hui AC (2002) Interobserver variations in intra-articular evaluation during arthroscopy of the knee. J Bone Joint Surg Br 84: Jomha NM, Borton DC, Clingeleffer AJ, Pinczewski LA (1999) Long-term osteoarthritic changes in anterior cruciate ligament reconstructed knees. Clin Orthop 358: Keene G, Bickerstaff D, Paterson R (1993) The natural history of meniscal tears in anterior cruciate insufficiency. Am J Sports Med 21: Larsen E, Jensen PK, Jensen PR (1999) Long-term outcome of knee and ankle injuries in elite football. Scand J Med Sci Sports 9: McDaniel WJ, Dameron TB (1983) The untreated anterior cruciate ligament rupture. Clin Orthop 172: Murrell GAC, Maddali S, Horovitz L, Oakley SP, Warren RF (2001) The effects of time course after anterior cruciate ligament injury in correlation with meniscal and cartilage loss. Am J Sports Med 29: Nikolic DK (1998) Lateral meniscal tears and their evolution in acute injuries of the anterior cruciate ligament of the knee. Arthroscopic analysis. Knee Surg Sports Traumatol Arthrosc 6: Paletta GA, Levine DS, O Brien SJ, Wickiewicz TL, Warren RF (1992) Patterns of meniscal injury associated with acute anterior cruciate ligament injury in skiers. Am J Sports Med 20:

9 Roos H, Lindberg H, Gardsell P, Lohmander LS, Wingstrand H (1994) The prevalence of gonarthrosis and its relation to meniscectomy in former soccer players. Am J Sports Med 22: Roos H, Adalberth T, Dahlberg L, Lohmander LS (1995) Osteoarthritis of the knee after injury to the anterior cruciate ligament or meniscus: the influence of time and age. Osteoarthritis Cartilage 3: Segawa H, Omori G, Koga Y (2001) Long-term results of non-operative treatment of anterior cruciate ligament injury. Knee 8: Smith JP 3rd, Barrett GR (2001) Medial and lateral meniscal tear patterns in anterior cruciate ligament-deficient knees. A prospective analysis of 575 tears. Am J Sports Med 29: Van Kampen A, Waal-Malefijit MC, Jersoch J, Castro WMH, Busch M, Pape M (1998) Interobserver variance in diagnostic arthroscopy of the knee. Knee Surg Sports Traumatol Arthrosc 6:16 20